1. Field of the Invention
The present invention relates to the investigation of subsurface earth formations, and more particularly to methods for determining one or more characteristics of an earth formation using a borehole logging tool.
2. Description of the Related Art
When drilling an oil and gas well, it is often desirable to run a logging while drilling (LWD) tool in-line with the drill string to gather information about the subsurface formations while the well is being drilled. The LWD tool enables the operators to measure one or more characteristics of the formation around the circumference of the borehole. Data from around the borehole can be used to produce an image log that provides the operator an “image” of the circumference of the borehole with respect to the one or more formation characteristics. The data can also be accumulated to produce a value of the one or more formation characteristics that is representative of the borehole circumference.
One type of LWD tool incorporates gamma-gamma density sampling to determine one or more formation characteristics. In gamma-gamma sampling, gamma rays are emitted from a source at the tool and scatter into the formation. Some portion of the radiation is reflected back to the tool and measured by one or more detectors. Formation characteristics, including the formation density and a lithology indicator such as photoelectric energy (Pe), can be inferred from the rate at which reflected gamma radiation is detected. Generally, the more radiation detected by the detectors the lower the density of the formation.
The amount of radiation detected is measured in counts, and is usually expressed in counts per unit time, or count rate. The statistical precision of the count rate is a function of the total counts acquired in a measurement. Precise measurements of low count rates require a longer acquisition time than equally precise measurements of high count rates. Generally, a measurement period of between 10 and 20 seconds is required to obtain a sufficient amount of data for a precise measurement of a formation characteristic. However, typical drilling rates require that the rotational period of the drill string, onto which the LWD tool is mounted, be less than one second. Thus, count rate data from several rotations must be combined to achieve a precise measurement.
In ideal conditions, the counts collected from the several rotations can be summed linearly. Many factors affect the accuracy of the measured count rate both at different points around the circumference of the borehole and at the same point from rotation to rotation. Therefore, various methods have been developed to account for the inaccuracy in the count rates as they are built up for several rotations. The effectiveness of such methods ultimately affects the accuracy of the assessment of the one or more formation characteristics.
One factor that affects the accuracy of the count rate data accumulated during the measurement period is the proximity of the detector to the borehole wall, or standoff. The standoff of the tool can vary azimuthally around the circumference of the borehole, as well as at the same point from rotation to rotation. When the standoff is low, and the detector is close to the borehole wall, the detector is reading radiation reflected primarily from the formation. When the standoff is high, drilling mud that is continually being circulated about the tool fills the annular space between the detector and the borehole wall. The detector in this case is then reading radiation reflected from the formation and the drilling mud, and the resultant count rate is not representative of the formation.
Typically, if the borehole is in gauge and of uniform circular cross-section, the standoff will be substantially consistent around the circumference of the borehole. With consistent standoff or small variations in standoff, known statistical methods can make adequate compensation for the effect of the drilling mud. However, many situations arise where the standoff can vary substantially for different azimuthal angles. More substantial variations in standoff impact the accuracy of the count rate and are more difficult to compensate, particularly as the offset becomes large. For example, the borehole gauge can be elliptical, and if the tool remains centered in the bore the standoff would be the greatest at the major axis of the ellipse. Thus, the mud would have a greater affect on the count rate when the detector is near the major axis, and a lesser affect on the count rate when the detector is near the minor axis. In another example, the gauge of the borehole can be oversized, though circular, elliptical, or otherwise. In such a situation, the tool may walk around the borehole tending to contact the borehole wall at many different points. In a borehole that is highly deviated or almost horizontal, the tool may sometimes climb the sidewalls. Irregular variations that occur when the tool walks in the borehole are difficult to compensate, especially when the standoff changes are large.
Another factor that must be accounted for, particularly when a formation characteristic representative of the borehole circumference is desired, is the variation in the measured parameter at different points around the circumference of the borehole. Typically, earth formations are sedimentary, and thus consist of generally homogenous horizontal layers. Occasionally, however, the layers will have discontinuities of notably different characteristics. The borehole may intersect the discontinuity such that a portion of the borehole circumference has different characteristics than the remainder. Even without a discontinuity, the characteristics of the borehole may be different in different portions of the circumference. For example, a highly deviated borehole may cross a horizontal boundary from one formation to the next at an angle. In some cases, a portion of the borehole circumference is representative of one formation while the remainder is representative of another formation. Such variations in formation characteristics can usually be seen in an image log.
Known techniques that attempt to compensate for perturbations in the count rate have tended to concentrate on achieving an accurate representative value of the formation characteristic for the borehole circumference, rather than an accurate borehole image. As such, the known techniques have relied on generalizations of the data in their methods. For example, U.S. Pat. No. 5,397,893 to Minette, discloses a method that groups or bins data by azimuthal angle, preferably by quadrant, or by the amount of standoff when the measurement is taken. The data that is grouped by azimuthal angle, that is the most useful for determining a borehole image, does not take in to account actual standoff. The data grouped by standoff is not associated with azimuthal angle to enable correlation with its position in the borehole.
Another system disclosed in U.S. Pat. No. 5,473,158 to Holenka et al. teaches a method whereby data is also grouped by quadrant. The statistical distribution of each quadrant is analyzed, and an error factor for each quadrant is calculated. The error factor is then applied to the entire quadrant, rather than the individual data grouped therein. Such generalization by quadrant is not ideal for devising a borehole image nor a representative formation characteristic of the borehole.
Therefore, there is a need for a method of measuring one or more characteristics of formation that more accurately accounts for perturbations in the measurements. Further, it is desirable that this method enable accurate imaging of the entire circumference of the borehole.